Microcalorimetry: A Novel Application to Measure In Vitro Phage Susceptibility of Staphylococcus aureus in Human Serum

Abstract

Infections involving antibiotic resistant Staphylococcus aureus (S. aureus) represent a major challenge to successful treatment. Further, although bacteriophages (phages) could be an alternative to antibiotics, there exists a lack of correlation in phage susceptibility results between conventional in vitro and in vivo assays. This discrepancy may hinder the potential implementation of bacteriophage therapy. In this study, the susceptibility of twelve S. aureus strains to three commercial phage cocktails and two single phages was assessed. These S. aureus strains (including ten clinical isolates, five of which were methicillin-resistant) were compared using four assays: the spot test, efficiency of plating (EOP), the optical density assay (all in culture media) and microcalorimetry in human serum. In the spot test, EOP and optical density assay, all cocktails and single phages lysed both methicillin susceptible and methicillin resistant S. aureus strains. However, there was an absence of phage-mediated lysis in high concentrations of human serum as measured using microcalorimetry. As this microcalorimetry-based assay more closely resembles in vivo conditions, we propose that microcalorimetry could be included as a useful addition to conventional assays, thereby facilitating more accurate predictions of the in vivo susceptibility of S. aureus to phages during phage selection for therapeutic purposes.

Keywords: MRSA; Staphylococcus aureus; bacteriophage therapy; human serum; microcalorimetry; susceptibility testing.

Figure 1

Circular visualization of the genomes of S. aureus strains in our panel compared to 205 S. aureus strains representing global genetic diversity. The maximum-likelihood tree in the figure describes the core SNP differences. From the inner to the outer circle, the first circle represents public genomes of S. aureus including the genomes from this study marked in blue and larger font, the second and third circles represents sequence type (ST) characterization visualized by color and text and the fourth circle represents CC information. Genomes without an ST and CC identification were symbolized with NA (Non-Available).

Figure 2

Presence of phages in three commercially available phage cocktails. Y-axis: Percentage of viral sequencing reads for the Russian- and Georgian-Pyofag cocktail (RPC and GPC, respectively) and the Intestifag cocktail (INT) mapped against the genome sequences of species types representing five genera that mostly comprise virulent phages against S. aureus.

Figure 3

Susceptibility of S. aureus isolates to phage cocktails as determined by OD assay in TSB. The OD of five MSSA strains, five MRSA strains and two laboratory strains (LS) inoculated with the (A) RPC, (B) GPC and (C) INT at different multiplicity of infection (MOI) was measured for 24 h. The OD over time (24 h) is depicted as percentage of area under the curve (AUC) compared to the growth control, which was set at 100%. All conditions were repeated in three independent experiments.

Figure 4

Metabolic activity of two phage sensitive strains in heat-inactivated human serum. Two phage-sensitive clinical S. aureus strains, Mup15 and Mup2723, were inoculated with phage cocktails RPC, GPC or INT an MOI 0.03 or with rifampicin (RIF; 40 µg/mL) or flucloxacillin (FLX; 128 µg/mL). As a positive control for bacterial growth, a control with PBS instead of phages or antibiotics was used. The heat flow was measured for 20 h and is shown as a percentage of the area under the curve (AUC) when compared to the positive growth control, which was set at 100%. All conditions were tested in three independent experiments.

Figure 5

Phylogenetic tree of single phages and reference strains. A maximum-likelihood phylogenetic tree was inferred for the RPCSa1 and RPCSa2 and reference strains representing the diversity of staphylococci phages using the tail sheath protein gene. The tree was mid-point rooted and bootstrap values are shown at the nodes. Scale bars show the number of nucleotide substitutions per site.

Figure 6

Susceptibility of S. aureus isolates to isolated single phages as determined by OD assay in TSB. The OD was measured for five MSSA strains, five MRSA strains and two laboratory strains (LS) inoculated with (A) RPCSa1 and (B) RPCSa2 at different MOIs. The OD over time (20 h) is shown as percentage of the area under the curve (AUC) compared to the growth control, which was set at 100%. ^# For clarity, values above 200% are not shown. All conditions were tested in three independent experiments.

Figure 7

Phage susceptibility in TSB compared to 82% human serum. Microcalorimetry was used to measure bacterial heat production after incubation of isolated single phages with two highly sensitive strains (Mup15 and SA2704), two moderate sensitive strains (MW2 and Mup3199) and two non-sensitive strains (Mu50 and Rww146) of S. aureus. Heat flow was measured for twenty hours in either (A) TSB or (B) 82% human serum and is shown as a percentage of the area under the curve of the growth control (no phages added). ^# For clarity, values above 200% are not shown. All conditions were tested in three independent experiments.